Key Features
Tech Stack
A concrete eval method: run it on a file that caused trouble in the past, then on the same file after patch/refactor. Before vs. after tends to be clearer than looking at scores in isolation.
Tech notes: multi-language via tree-sitter (Python, JS/TS, Java, Go, Rust, C/C++, etc.), 100% local (no telemetry / no external APIs), deterministic (no LLMs for the scores).
Happy to answer questions about the metrics.
macOS / Linux
docker run -it --rm \ -e MISTSEEKER_LICENSE_KEY=YOUR_LICENSE_KEY \ -v "$(pwd)":/app/code \ -v "$(pwd)/reports":/app/out \ tongro2025/mistseeker:latest \ analyze --input-root /app/code \ --output-json /app/out/mistseeker_report.json \ --output-image /app/out/mistseeker_report.png
Windows (PowerShell)
docker run -it --rm ` -e MISTSEEKER_LICENSE_KEY=YOUR_LICENSE_KEY ` -v "${PWD}:/app/code" ` -v "${PWD}/reports:/app/out" ` tongro2025/mistseeker:latest ` analyze --input-root /app/code ` --output-json /app/out/mistseeker_report.json ` --output-image /app/out/mistseeker_report.png
Windows (CMD)
docker run -it --rm ^ -e MISTSEEKER_LICENSE_KEY=YOUR_LICENSE_KEY ^ -v "%CD%:/app/code" ^ -v "%CD%/reports:/app/out" ^ tongro2025/mistseeker:latest ^ analyze --input-root /app/code ^ --output-json /app/out/mistseeker_report.json ^ --output-image /app/out/mistseeker_report.png
Not affiliated with Hacker News or Y Combinator. We simply enrich the public API with analytics.